Add the dbt Cloud capability

After you have created a dbt connection, you have to add the capability to the connection.

Required permissions

Steps

  1. Open a site.
    1. On the main toolbar, click Products icon → Cogwheel icon Settings.
      The Settings page opens.
    2. In the tab pane, click Edge.
      The Sites tab opens and shows a table with an overview of your sites.
    3. In the table, click the name of the site whose status is Healthy.
      The site page opens.
  2. In the Capabilities section, click Add capability.
    The Add capability page appears.
  3. Select the Technical Lineage for dbt Cloud capability template.
    Important Do not select the Technical Lineage for dbt (Cloud) capability template. That one is used for the dbt Core integration via Cloud Storage connection.
  4. Enter the required information.
    FieldDescriptionRequired?

    Name

    The name of the capability.

    Yes

    Description

    The description of the capability.

    No

    Source ID

    The name of the data source. The name must be unique and cannot contain special characters, for example, /.

    Yes

    TechLin Admin Connection (in preview)

    If you want to use the OAuth authentication type to connect to the Collibra Data Lineage service instances, you have to create a Technical Lineage Admin Edge or Collibra Cloud site connection and select the OAuth authentication type. Then, in this field, you specify the name of the Technical Lineage Admin Edge or Collibra Cloud site connection.

    No

    dbt Connection

    The dbt connection that you created.

    Yes

    Environment Ids

    The IDs of the environments that Collibra Data Lineage uses to download job artifacts.

    Enter an array of environment IDs, for example [123456, 987654]. This field is required if you do not enter a value for the Admin URL field in the dbt connection.

    If you enter values for both the Admin URL and Environment Ids fields, the Environment Ids field takes precedence.

    No

    Source Configuration

    The source configuration to reduce the amount of data objects to be downloaded and enhance the performance of CollibraData Lineage in the following ways:

    • Filter the projects and jobs to be downloaded. Include projects and jobs to be downloaded by specifying the filter property.
    • Specify different Collibra system names for different projects by specifying the collibraSystemNames property .
    • Map a materialization as a view instead of a table by specifying the materializedMapping property.

    Specify the following properties in JSON format and enter the content in this field.

    Copy
    {
        "collibraSystemNames":{
            "projects":[
            {
                "project_id":654321,
                "collibraSystemName":"SystemName"
            }
           ]
        },
        "filter":{
            "jobIds":[1234],
            "projectIds":[654321]
        },
        "materializedMapping":{
            "ELS_MATERIALIZE_MULTIPLE_EXTERNAL_TABLES":"VIEW"
        }
    }

    Property

    		Description

    Required?

    collibraSystemNames

    You can use this section to specify the Collibra System Name for each project.

    No

    projects

    This section contains the project names and the Collibra system names.

    No

    project_id

    Your project ID. You can find the project ID in the dbt URL right after projects. For example, if your dbt URL is https://cloud.getdbt.com/develop/54321/projects/12345 , your project_id is 12345.

    No

    collibraSystemName

    The system or server name of the data source. This is also the name of your System asset in Data Catalog:

    Specify this property with the same name as the name of the System asset that you created when you registered the data source.

    In the following code example, the project with the 12345 project ID is stitched to the systemname1 System asset in Data Catalog. 

    {
   "collibraSystemNames":{
	"projects":[
	    {"project_id":"12345","collibraSystemName":"systemname1"}
	]
   },
}
    		No
    filter

    You can use this section to include projects and jobs to be downloaded. Collibra Data Lineage downloads and processes only the specified jobs and projects.

    In the following code example, the job with the 1234 job ID and the projects with the 98 and 5678 project IDs are downloaded. 

    {
       "filter": {
	   "jobIds": [
	       1234
	   ],
	   "projectIds": [
	       98,
	       5678
	   ]
	}
}

    No

    jobIds

    The job IDs of the jobs that you want to include.

    Specify an integer. Do not specify a string.

    To get your job ID, in your dbt, select Deploy and then Jobs. Select a job and you can find your job ID in the URL. For example, if your URL is cloud.getdbt.com/deploy/65432/projects/23456/jobs/123456, 123456 is your job ID.

    No

    projectIds

    The project IDs of the projects that you want to include.

    Specify an integer. Do not specify a string.

    You can find the project ID in the dbt URL right after projects. For example, if your dbt URL is https://cloud.getdbt.com/develop/54321/projects/12345 , your project_id is 12345.

    		No
    materializedMapping

    Indicates how materializations in dbt are mapped. If you do not specify this property, CollibraData Lineage maps materializations to tables by default. You can change the mapping of a materialization to view.

    In the following example, the ELS_MATERIALIZE_MULTIPLE_EXTERNAL_TABLES materialization is mapped to a view. 

    "materializedMapping":{
    "ELS_MATERIALIZE_MULTIPLE_EXTERNAL_TABLES":"VIEW"
}
    		No
    Tip If you previously created a technical lineage for this data source with connection definitions by using the lineage harvester, you can enter the content from the <sourceId>.conf file in this field.

    No

    Property

    This section contains the custom parameters you can specify to create technical lineage. Click Add property to add a property.

    Warning If you are a Collibra Platform for Government customer, this field is required to connect to a Collibra Data Lineage service instance:
    TypeValue TypeNameValue

    Text

    Plaintext

    techlinHost

    This is the URL of the Collibra Data Lineage service instance to which you want to upload metadata, for example techlin-europe-west1.collibra.com.

    		Example: techlin-europe-west1.collibra.com

    Text

    Secret

    techlinKey

    This is the unique API key to connect to a Collibra Data Lineage service instance.

    Specify a unique user key for each Collibra environment. If you're not sure what your user key is, contact your Collibra Collibra Account Team.

    		<your-techlin-key>

    Yes for US government customers.

    Dependent On Sources

    This option allows you to provide table-definition details from an independent data source to a data source that is dependent on those details. This is needed to avoid analysis errors and to have a complete lineage that includes lineage from the SQL statements from dependent data sources.

    To use this option, enter the source ID of the independent source.

    Let's consider an example. Let's say that you want to create a technical lineage for two data sources:

    Database1 contains the DDL that specifies that the database has a table named "Table1", which has three columns: Col1, Col2, Col3, and Col4.

    Database2 contains an SQL statement: SELECT * from Database1.Schema1.Table1.

    The SQL statement in Database2 refers to the table in Database1. Therefore, to get lineage from the statement in Database2, the table definition from Database1 must be known. In this case, we say that Database2 is dependent on Database1. Database1 is considered the independent data source.

    To configure this option, specify the Source ID of the independent data source, in this example, Database1, as shown in the following imagee

    An image showing the Dependent on Sources field in the Edge capability

    Important If a dependent data source contains lowercase column names, this feature will only work for the following dialects: Oracle, Snowflake, and Teradata. For all other dialects:
    • An analyze error is raised, prompting you to provide the DDL file.
    • The only workaround is to consolidate your SQL statements and DDL file in a single data source.

    For complete information, go to Sharing database models across data sources.

    		 

    Delete Raw Metadata After Processing

    Technical lineage via Edge harvests raw metadata from specified data sources and uploads it in a ZIP file to a Collibra Data Lineage service instance. This option indicates whether the raw metadata should be deleted from the Collibra Data Lineage service instance after the metadata that is targeted for ingestion in Data Catalog is processed.

    Select this option to indicate that the raw source metadata is deleted after processing.

    Clear the checkbox to keep the raw source metadata after processing. In this case, it is stored in the Collibra infrastructure.

    No

    Analyze Only (Deprecated)

    Important This option is deprecated and will be removed in a future version of Collibra. We recommend that you no longer use it. The mandatory Processing Level setting, below, replaces this option.
    • The "Analyze" option in the Processing Level setting is the equivalent of selecting the Analyze Only option.
    • The "Sync" option in the Processing Level setting is the equivalent of clearing the Analyze Only option.

    No

    Processing Level

    Important This setting replaces the deprecated Analyze Only option, which will be removed in a future version of Collibra.

    For each of your data sources, you have to specify one of the following values: Load, Analyze, or Sync. Then, when you synchronize your technical lineage, the following process begins:

    1. Metadata for all data sources is loaded, regardless of the value of this setting for a particular data source.
    2. Metadata from data sources for which the value of this setting is either Analyze or Sync, is analyzed.
    3. Metadata from data sources for which the value of this setting is Sync, is synchronized.

    ValueDescription
    Load

    Harvest metadata from the data source and upload it to your Collibra environment. This allows you to inspect and, if necessary, edit the harvested metadata before uploading it to the Collibra Data Lineage service instance for analysis.

    When the job is done, you can download and review the metadata:

    1. Open the Activities list.
    2. In the row containing the job, click Result.
      The Synchronization Results dialog box appears.
    3. Click download and save the ZIP file to your hard drive.

    Tip The download link resembles the following: https://integrations.collibra-abc.com/rest/2.0/files/01944f12-7665-7d9c-8bc5-aa426b6a63cc. Take note of the file ID, in this example: 01944f12-7665-7d9c-8bc5-aa426b6a63cc. After you inspect the metadata, you can send the ZIP file for analysis by using the "Analyze files" option. Alternatively, you can upload the ZIP file using the POST /files API. In either case, you need to specify the file ID.

    Analyze

    Load and analyze the metadata on the Collibra Data Lineage service instance.

    Synchronization does not start after analysis; it starts only after either:

    Important  If you want to synchronize multiple data sources, we strongly recommend that you select this option in the respective Edge or Collibra Cloud site capabilities for each of your data sources. This allows you to synchronize all data sources in a single job, thereby maximizing efficiency and mitigating the risk of failed synchronization jobs.
    Sync

    Load, analyze, and synchronize metadata from all data sources. Synchronization starts – or is queued, if another synchronization job is running – immediately after analysis.

    Important If you want to synchronize multiple data sources and you select this option, each data source is processed as a separate job. This is highly inefficient and will likely lead to failed sync jobs. For complete information and important considerations, go to Tips for successful lineage synchronization.

    Yes

    Active

    The option determines whether to include or remove the technical lineage of the data source.

    Select this option to include the technical lineage of this data source.

    Clear the checkbox to exclude the technical lineage of this data source.

    Yes

    Debug

    This setting is not valid for this integration. It should be set to false.

    No

    Log level

    Only complete this field on the request of or together with Collibra Support.

    No

  5. Click Create.
    The capability is added to the Edge or Collibra Cloud site.
    The fields become read-only.

What's next

You can now synchronize the technical lineage.